以长江、黄河源区为研究对象,应用大尺度半分布式水文模型(VIC),结合江河源区气象站多年实测温度、降水数据,检验了VIC模型的适用性。模型能较好模拟江河源区地表径流,其Nash系数和相关系数分别达到了0.8533和0.9302(长江源区),0.8892和0.9248(黄河源区)。基于率定后的VIC模型,运用高分辨率的动力降尺度气象强迫资料,分析了未来气候变化情景下,江河源区径流量可能变化趋势。结果表明,未来30~50a,长江、黄河源区年均径流量将分别增加8.58%、9.19%;未来80~100a,长江、黄河源区年均径流量将分别增加17.16%、7.21%。相对于2030~2049年而言,尽管年均降水增加0.06mm/d,但是黄河源区2080~2099年径流量却将减少1.98%。运用植被情景假设及2030~2049年动力降尺度气象资料,模拟分析了植被变化对江河源区地表径流的影响。从4种地表覆被情景假设可以看出,林地地表覆被产生径流量最小,裸地最大。
By taking the source region of the Yangtze River and the Yellow River as research object,the suitability of large-scale semi-distributed hydrologic model(Variable Infiltration Capacity,VIC)was tested by using observed daily maximum ground temperature,daily minimum ground temperature and daily precipitation data of the source region of the Yangtze River,the Yellow River and their surrounding in 1980~ 1999.The results showed that VIC model was suitable to simulate the runoff of the source region of the Yangtze River and the Yellow River.The NASH coefficient and correlation coefficient of VIC in the source region of the Yangtze River was 0.853 3and 0.930 2,respectively,while that in the source region of the Yellow River was 0.889 3and 0.924 8,respectively.Then by using the validated large-scale distributed hy-drology model and high resolution dynamic downscaling meteorological forcing data,the possible trends in runoff of the source region of the Yangtze River and the Yellow River were analyzed under future climate warming.The average annual runoff of the source region of the Yangtze River and the Yellow River will increase by 8.58%and 9.19%in the future 30to 50years,respectively.In the future 80to 100years,it will increase by 17.16%and 7.21%.And relative to the annual precipitation of 2030~2049,the annual precipitation of 2080~2099will increase up 0.06mm/day,but the average annual runoff of the source of the Yellow River will reduce only by 1.98%in the next 50years.The variations of runoff in the source area of the Yangtze River and the Yellow River were analyzed by using the high resolution dynamic downscaling as meteorological input data for the future 30to 50years,based on four scenario simulations of the land use/ cover change.The impact of different vegetation cover type on runoff was compared.These results indicate that the runoff is the minimum at forest land,but that is the maximum at bare land.